Advancing Large Multi-modal Models with Explicit Chain-of-Reasoning and Visual Question Generation

TL;DR

This paper introduces a novel vision-language model that employs explicit chain-of-reasoning and visual question generation, improving interpretability and robustness in visual content understanding.

Contribution

The paper develops a new dataset and training approach enabling VLMs to generate questions and perform iterative reasoning for enhanced interpretability.

Findings

Improved reasoning accuracy in VLMs

Enhanced robustness through question-asking mechanism

Demonstrated effectiveness on diverse visual tasks

Abstract

The increasing demand for intelligent systems capable of interpreting and reasoning about visual content requires the development of large Vision-and-Language Models (VLMs) that are not only accurate but also have explicit reasoning capabilities. This paper presents a novel approach to develop a VLM with the ability to conduct explicit reasoning based on visual content and textual instructions. We introduce a system that can ask a question to acquire necessary knowledge, thereby enhancing the robustness and explicability of the reasoning process. To this end, we developed a novel dataset generated by a Large Language Model (LLM), designed to promote chain-of-thought reasoning combined with a question-asking mechanism. The dataset covers a range of tasks, from common ones like caption generation to specialized VQA tasks that require expert knowledge. Furthermore, using the dataset we created, we fine-tuned an existing VLM. This training enabled the models to generate questions and perform iterative reasoning during inference. The results demonstrated a stride toward a more robust, accurate, and interpretable VLM, capable of reasoning explicitly and seeking information proactively when confronted with ambiguous visual input.